Melanoma of the uveal tract is the most common primary intraocular disease in adults that can be fatal. Its clinicobiological course, such as the long latency and rare spontaneous regression, suggests a host defense mechanism. Indeed, induction of tumor-specific T-cells in animal and in clinical trials have been documented. Recent clinical trials of adoptive transfer of in vivo expanded cytotoxic T-cells (CTLs) and active specific immunotherapy strongly suggest that the CTL may be pivotal for tumor rejection in human. We have identified CTLs with different characteristics in the peripheral blood 80% of choroidal melanoma patients studied. In this proposal CTL will be generated from patients with active disease from the peripheral blood and from tumor tissues. The peripheral blood is a renewable source of lymphocytes, an important consideration if therapy is ever contemplated. On the other hand, lymphocytes in tumors maybe preselected in vivo for antitumor activity. Information regarding CTLs from both tissue compartments are complementary and necessary for additional insight into the host-tumor interaction. (1) Peripheral blood lymphocyte. Because the expansion of particular populations of CTLs depends in large part on the stimulus used, peripheral blood mononuclear cells will be stimulated with antiCD3 monoclonal antibodies to capture all T-cell reactivities as well as with mixed lymphocyte tumor cell cultures (MLTC) to generate both MHC- unrestricted and MHC-restricted CTLs. The latter CTLs will be stimulated by a "simulated autologous" allogeneic choroidal melanoma cell line matched in the HLA-A region. All CTLs will be characterized by their cluster determinant (CD) phenotype, T-cell receptor (TcR) chains and specificity of lysis by cold target competition. Existence of CTLs with specificities predicted from bulk cultures will be proven at the clonal level. TcR gene usage in bulk culture will demonstrate oligoclonal proliferation of CTLs, and therefore, limited recognition of melanoma-associated antigens. TcR gene usage will be compared in T-cell clones with similar reactivities to determine if they recognize the same epitope. The role of MHC Class l and ll molecules and other accessory/adhesion molecules in T-cell killing of melanoma cells will be studied by blocking with the appropriate monoclonal antibodies. The MHC-unrestricted CTLs will be examined most closely. (2) Tumor-infiltrating lymphocytes. Immunostaining will be used to identify the CD phenotypes Of the T-cell subsets in primary lesions and the cytokines released at the tumor site. CTLs will also be released from tumor tissues. They will be characterized in bulk culture and at the clonal level. CTL derived from peripheral blood and in situ of the same individual will be compared. In summary, these studies hope to identify the role(s) of antitumor CTLs in tumor rejection in choroidal melanoma patients. Furthermore, information regarding the melanoma-associated antigens recognized may be useful for designing immunotherapeutic protocols for this disease.